As cancer involves proliferation of altered cell types that produce high levels of specific proteins and some enzymes, such as proteases, it will not only modify the array of existing blood proteins (the "proteome") but also their metabolic products, i.e., peptides (the "peptidome"). We have developed and shown proof-of-principle of a prototype technology platform for automated, magnetic bead-based, solid-phase extraction of peptides from microliter volumes of serum, coupled to a MALDI-TOF mass spectrometric (MS) read-out. We have used this system to analyze different types of cancer and found that selected peptides, when combined as signatures, have utility as surrogate biomarkers for cancer-specific exoprotease activities, or panels of activities. Serum peptide profiling is thus a form of activity-based proteomics, i.e., monitoring blood proteome "metabolomic" products, and the degradative patterns appear to hold information for detection and classification of cancer. This proposal applies specific expertise from well-established clinical and research groups at Memorial Sloan-Kettering Cancer Center (MSKCC) and at New York University (NYU) to assess our analytical platform and apply it to detect prostate and breast cancer-specific patterns. Unlike earlier studies, we will leverage the clinical resources of MSKCC to analyze a far larger number of samples than previously attempted. We propose to assess robustness and reproducibility of both system and approach using mouse models, clinical samples from prostate cancer and breast cancer patients, and samples from healthy volunteers. Our proposal includes the following initiatives: (1) establish a mirror site at NYU of primary platform A (robotics plus MALDI-TOF MS) in its entirety, followed by comparative performance and reproducibility assessment; (2) develop and assess a secondary platform B (LC-MS); (3) develop and evaluate an alternative, MS-based functional proteomics platform to assay cancer-specific protease panels in plasma; (4) introduce quantitative components (using a repository of non-degradable, isotopically-labeled reference peptides) to both platforms A and B for direct measurements or for assays; (5) develop robust and user-friendly support technologies/programs for signal processing, data storage, analysis and sharing; (6) establish a repository of 1,400 reference samples and 120 reference and assay-substrate peptides. Relevance: Our proposal will evaluate and document whether serum peptide patterns have diagnostic value for cancer detection, or may mark a given clinical outcome, or may distinguish clinically insignificant from significant cancer. Such a test could, for example, identify patients with newly diagnosed cancer who might safely avoid surgery or radiation. [unreadable] [unreadable] [unreadable]